Article Text


Churg-Strauss vasculitis syndrome and leukotriene receptor antagonists
  1. J Ng,
  2. R Savage,
  3. F McQueen
  1. Department of Rheumatology, Auckland District Health Board, Auckland, New Zealand and the Centre for Pharmacovigilance, Dunedin School of Medicine, University of Otago, New Zealand
  1. Correspondence to:
    Dr F McQueen
    Department of Rheumatology, Building 7, Auckland Hospital, Private Bag 92024, Auckland, New Zealand;

Statistics from

A 28 year old Middle Eastern woman presented to hospital with shortness of breath and new onset numbness of her foot. Five years previously, in Iraq, she had been diagnosed with severe brittle asthma and sinusitis. After 2 years of management with oral (up to 15 mg prednisone) and inhaled steroid, zafirlukast, a leukotriene receptor antagonist (LRA) was added to her treatment regimen. Despite this, her asthma remained poorly controlled and zafirlukast was stopped after 12 months. Subsequent clinical improvement allowed the prednisone dose to be tapered and then stabilised at 7.5 mg/day. By the time she migrated to New Zealand she had been receiving that dose for 1 year.

At presentation to Auckland City Hospital she was clinically in left ventricular failure with a raised jugular venous pressure. Neurological examination disclosed features of a common peroneal nerve palsy, presumed to be a manifestation of mononeuritis multiplex. Investigations showed a leucocytosis with extreme eosinophilia (white cell count 28.3×109/l, eosinophils 13.3×109/l) and erythrocyte sedimentation rate of 40 mm/1st h. Renal function and liver function were normal. Troponin T was raised at 0.64 μg/l (normal <0.03 μg/l), indicating myocyte injury. She was antineutrophil cytoplasmic antibody and antinuclear antibody negative. There was moderate left ventricular impairment on echocardiogram, with an ejection fraction of 45% and a small pericardial effusion. Chest radiography showed cardiomegaly and a fine nodular pattern at the lung bases, which resolved after starting treatment. Attempts to confirm vasculitis on sural nerve and muscle biopsies were unsuccessful.

This patient fulfilled American College of Rheumatology criteria for a diagnosis of Churg-Strauss syndrome (CSS).1 As there was evidence of cardiac injury, she scored 1 on the five factors score.2

She was treated with prednisone initially at 1 mg/kg (later tapered to 15 mg/day) and given monthly pulse cyclophosphamide at 0.5 mg/m2, to continue for 12 months according to the protocol proposed by Guillevin and Pagnoux.3 She has responded well symptomatically with normalisation of the eosinophil count and erythrocyte sedimentation rate. On echocardiography, the ejection fraction remains unchanged but the pericardial effusion has resolved.


Our patient’s experience raises several interesting points. To our knowledge, this is the first case reported in Australasia of CSS occurring in a patient who had received LRA treatment. A possible causal association between the use of zafirlukast and CSS was first suggested in 1998.4 The New Zealand Intensive Medicines Monitoring Programme has prospectively monitored adverse events in 1535 patients who received prescriptions for montelukast from 1999 to the present.5 Forty three adverse events were considered associated with montelukast, but these did not include CSS or related features. The annual incidence of CSS in patients with asthma has been estimated at 64.4 per million patients.6 Thus, any link between montelukast and CSS may not have been detected owing to the small size of the cohort as LRA have not been widely used in New Zealand.

In most reports of CSS occurring in patients receiving LRA, extrapulmonary vasculitic features have developed after steroid reduction.7,8 It has been proposed that this may be because vasculitis has been “unmasked” by reduction of previously suppressive doses of steroid.7 This would not apply in our patient as she developed features of CSS after receiving a stable dose of 7.5 mg of prednisone for 12 months.

Other explanations for the association between treatment with an LRA and CSS in this patient include the possibility that zafirlukast has an immunomodulatory role in promoting the development of vasculitis directly (albeit some time later), as has been suggested by Guilpain et al.9 Alternatively, the association may have been purely coincidental, and the fact that zafirlukast was stopped 18 months before presentation with CSS makes it harder to implicate the drug in disease pathogenesis in this case.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.